Comments on Dataset Exchange Use Cases and Requirements FPWD

Dear Colleagues,

Apologies for being slightly late, but I have been travelling and have a lot of email to catch up.

1. Use case 5.32 Relationships. 
I suggest adding an example of the 'Update' relationship alongside 'Aggregation', 'Derivation' and 'Citation'. From my point of view, this is not quite the same as version (or maybe it is?). I work in weather forecasting, where information, and the underlying datasets are routinely updated, whether because more new data is available, a better simulation is available, or errors discovered, or too much time has elapsed and later is better quality. The updated dataset is a complete replacement, and generally the original dataset should be made inaccessible. 

It is not clear to me how the extracted requirements best address this use case. It is probably related to '5.38 Time related aspects' and possibly '5.47 Define update method'.

2. Use Case 5.27 Modelling Temporal Coverage
A list of Met Ocean examples are given of "temporal properties of ensemble forecast data:
    simulation event time
    analysis time (aka run time or reference time)
    assimilation window
    datum time
    forecast computation time
    validity time
    partial forecast time
    re-analysis event time
    forecast model run collection time"

Datasets that are include an ensemble of values, or single ensemble members, as proxies for a probability distribution, are an interesting use case to be considered, but generally they are orthogonal to temporal aspects. The examples above generally refer to any meteorological forecast data, including ensembles

Of the examples given:
    simulation event time [Not sure what is meant]
    analysis time (aka run time or reference time) [OK]
    assimilation window [OK]
    datum time [Not sure what is meant]
    forecast computation time [not sure what is meant, but probably the time taken to compute a forecast, rather than the time when the computation starts, or when the forecast finishes other wise known as the result time]
    validity time [OK]
    partial forecast time [Not sure what this is]
    re-analysis event time {Not sure what this is either]
    forecast model run collection time [Not sure what is meant, could be 'data cut-off time', when data ingest stops and forecast simulation starts, or when the finished forecast products are available for distribution]

I am happy to supply a list of meteorological commonly recognised list of example times, with explanations, if needed.

Chris

Chris Little
Chair, OGC Meteorology & Oceanography Domain Working Group
Member OGC Architecture Board

IT Fellow - Operational Infrastructures
Met Office  FitzRoy Road  Exeter  Devon  EX1 3PB  United Kingdom
Tel: +44(0)1392 886278  Fax: +44(0)1392 885681  Mobile: +44(0)7753 880514
E-mail: chris.little@metoffice.gov.uk  http://www.metoffice.gov.uk

I am normally at work Tuesday, Wednesday and Thursday each week

Received on Wednesday, 24 January 2018 14:25:57 UTC